开云体育

This is my review

Thanks, added to /g/nanovna-users/wiki#NanoVNA-H4

My 2.8" NanoVNA-H v3.4 arrived. I ordered it ($50usd) from hugen's Alibaba store on December 31st 2019 and it arrived January 17, 2020 via parcel post. As luck would have it, the 4" NanoVNA-H4 was announced a few days after my order and the remainder of my Christmas money went towards purchasing it on January 10, 2020. Maggie King, at hugen's Alibaba store, handled both transactions and was a pleasure to deal with.

This is my review of the 2.8" NanoVNA-H v3.4 for fellow members. I was waiting for erik's review but when my unit arrived I figured I'd take a shot at it.

What's in the box.

The 2.8" NanoVNA-H v3.4 came in well packaged storage box with molded compartments for the NanoVNA-H and accessories. The storage box was packed in a shipping container with sufficient padding to survive most in-route mishaps. If you order a NanoVNA-H v3.4 elsewhere and its not pictured in an identical storage box, your spider sense should immediately start tingling.

The contents of the storage box included:
NanoVNA-H (with 650mAh battery)
USB Type-C data cable x1
20cm SMA male to male RG174 RF cable x2
SMA male calibration kit - OPEN x1
SMA male calibration kit - SHORT x1
SMA male calibration kit - LOAD x1
SMA female to female connector x1
Type-c to Type-c cable x1

Visual improvements over original NanoVNA.

1. hugen's original NanoVNA was an open case design. Surprising the open case design doesn't seem to hurt its rf performance, but it always appears to be one small screw dropping inside from the graveyard. The open case design also does not provide guards for the fragile switches on top of the unit. The NanoVNA-H v3.4 comes in a well fitted ABS case and should survive an accidental drop onto a carpeted floor ( tile or concrete, not so much) and provides guards for the top mounted switches.

2. hugen's original NanoVNA came with 400mAh battery with a 2 hr off-line operating life. The NanoVNA-H v3.4 comes with a 650mAh battery. The off-line operating life should be around 4 hrs. The original NanoVNA did not have the necessary diode installed (v3.3 up does) to provide battery voltage indication. This diode is installed in the NanoVNA-H v3.4. The larger battery and a new power chip decrease some of the noise that was limiting measurement range during usb connection.

3. A Type-c to Type-c cable was not part of hugen's original NanoVNA accessories primarily because the termination resistors that allow Type-c to Type-c data communications were not installed on the board. Purchasing a USB OTG cable usually worked. The termination resistors are already installed on the NanoVNA-H v3.4 (likewise the V3.3) so direct Type-c to Type-c cable connection is possible.

4. A label on the rear of the NanoVNA-H v3.4 uniquely identifies it with a QR-code and serial number.

5. Turning on the NanoVNA-H v3.4 presented a version screen of 0.4.4-1 10DEC19. hugen has just released firmware January 18th, 2020 0.4.5-4 and I upgraded to the large font version of it. I had previously upgraded my older NanoVNA-H v3.3 to the large font version and discovered there is a trade-off. You have to give up one of the five save locations to make room for the larger fonts. The trade-off is worth it to me.

6. The specified NanoVNA-H v3.4 frequency range is 50k-1.5GHz. The 0.4.4-1 firmware allows setting a lower frequency of 10kz and a high frequency of 1.5GHz. [Note per hugen: The lower bound frequency is limited by the IF frequency. It needs twice or more of IF frequency. The other limitation is the mixer coupling capacitor.] Also, erik has submitted a request to add a menu option to unlock the high frequency to 2.1 GHz for those who understand the cons (increase in noise). Maybe something like the clearconfig 1234 command ???.

7. The NanoVNA-H v3.4 shortens the two accessory RG174 cables from 30cm to 20cm to reduce insertion loss. It finally comes with a form of written documentation, a Menu Structure Map.

8. The slightly annoying "feature", where the power LED stays on for a few moments after switching the NanoVNA-H off, seems to have been eliminated in v3.4. If you are not sure if you really received a NanoVNA-H v3.4 and don't want to open the case, check to see if the power light goes off immediately after you switch it off.


So what's under the hood? …

1. All the following under the hood comparisons were made using NanoVNA-H v3.3 and NanoVNA-H v3.4 devices. Both devices were upgraded to hugen firmware version 0.4.5-4. The v3.3 measurement performance pretty much applies to all prior version NanoVNA's.

2. From hugen: "In order to make it easier for geeks to create more possibilities, UART1, PC14, and PC15 are introduced on the v3.4 PCB." So those members who have been clamoring for UART access in order to add Bluetooth capability, I can confirm that you've got your wish.

3. As a side-line, I purchase components for resell at swap meets and on eBay. Things like attenuators, filters, RF amplifiers, loads etc. I use the s21 function of the NanoVNA-H for the majority of my testing so its noise floor specification is important to me. If I can get at least a 40 dB noise floor for my pre-amp measurements then I'm usually satisfied.

The attached noise floor measurements were made with CH1 of both devices terminated with a 50 ohm load. The test ranges were 10 kHz - 1 GHz and 1-1.5 GHz. The NanoVNA-H v3.4 measurements show a marked improvement in noise floor, in some cases 20dB or greater, over the NanoVNA-H in both frequency ranges. Hugen said that 50dB down was going to be hard for him to guarantee in the 1 - 1.5 GHz range but my unit seems to be right there. All measurements were taken with the units connected to my pc so that I could quickly take screen shots (Tks Rune for the new NanoSaver-saver screen shot capability), that should also be the worst case configuration for noise measurements. No averaging was used, or available in the firmware I installed.

At 1 GHz, in the 10 kHz to 1 GHz scan, I measured 60dB down. At 1 GHz, in the 1 - 1.5 GHz scan, I was 50dB down. I've seen similar band switching noise floor rises in professional level equipment so I suspect something similar is occurring. Maybe the drive level gets re-adjusted when the range switch over is made.

4. After the noise floor measurements I wanted to see what some real word components looked like. I have a 1300 MHz BPF which was a perfect candidate for testing the 1-1.5 GHz range. The filter skirts were steeper and the bandpass was flatter for the NanoVNA-H v3.4.

I have an Agilent 70dB step attenuator (dc-4GHz) that I trust so I used it to see how far down I could go in the 1-1.5 GHz range. The NanoVNA-H v3.4 hit the limit at -40dB and the NanoVNA-H at -20 dB, both limits were about 10dB above their respective noise floors. The -40dB down level is the water level mark for usability in my measurements. I can see why erik was enthusiastic about opening up the stop frequency to 2.1 GHz, as long as the user is aware of the usual caveats.

The last measurement I made was a vswr antenna measurement on CH0. I purchased the antenna in a set of four at the last swap meet I attended for re-sell on eBay. Its a well crafted Japanese antenna marked for 1450 MHz. I had previously measured the antenna using my NanoVNA-H but couldn't get a useable indication. The NanoVNA-H 3.4 easily showed a VSWR of 1.1 and the antenna was well matched to 50 ohms.

Hugen and everyone who contributed to the v3.4 design are to be commended.


Hope this review helps someone in deciding on their next purchase. At the same price as the current NanoVNA-H is going for, the NanoVNA-H v3.4 was a no-brainer for me. I have a feeling the NanoVNA-H is reaching the limits of its design with v3.4. If I were buying today, I would wait for reviews of the 4" NanoVNA-H4 to come in before making my purchase. Gabriel's 3 GHz V2 hardware, firmware, and software still appear to be a month or more away from being finalized. The NanoVNA-F has announced plans to increase it's specifications to 1.5 GHz, so it is in the game too.

This message was re-posted with correct photos (Tks Andy)

- Herb

Ok, Erik!
Thank you very much!
73!

I was afraid shorting R13 would overdrive CH1 as I did not have the resistors to change the CH1 input (R29-R31)
Current CH1 headroom is less than 10dB
--
NanoVNA Wiki: /g/nanovna-users/wiki/home
NanoVNA Files: /g/nanovna-users/files
Erik, PD0EK

On Mon, Jan 20, 2020 at 11:45 AM, RFy wrote:

How fast full sweep and how many sweep points on this variants?

Same as the original nanovna.

hugen

On Sun, Jan 19, 2020 at 11:01 PM, Jos Stevens wrote:

You have to schort R13 and change? R22 upto included R24 also :

Jos.

I see it.
But Erik wrote only about R9, R29, R30, R31 for improving CH0.
Maybe he left R13 150 ohm? :-) That's why I asked him.

How fast full sweep and how many sweep points on this variants?

NanoVNA-H rev3.4 is fully compatible with previous versions of NanoVNA firmware.
The NanoVNA-H4 uses a Cortex-M4 MCU and a 4-inch LCD, so it is no longer compatible with previous firmware versions.

hugen

Hello all, I'm at a loss. I was able to update firmware by going into DFU mode on my nanoVNA, but now that is not working at all since firmware version 0.5.4. The DfuSe Demo (v3.0.6) is not recognizing the nanoVNA"c when I enter DFU mode either via "Config", "Reset and Enter DFU" or by putting the short across the pins to force DFU mode. My PC recognizes the difference between my virtual com port (normal mode) and Universal Serial Bus devices, STM32 BOOTLOADER mode (DFU) mode.

I'm not sure what has changed. Any troubleshooting tips would be greatly appreciated!

Thanks,
Brian

David,
Thank you.? I accidentally posted to the wrong list.? Sorry for the extra bandwidth.
Ray

--
They say a smart person learns from their mistakes. A
wise person learns from the mistakes of others.

Roger,
Have a look at Erik's post from earlier today. Those are the changes that are documented in the latest schematics for hugen's design v3.4
I hope you have a microscope and a steady hand in order to change those parts.?

RegardsLarry

On Fri, 17 Jan 2020 at 1:41 PM, Roger Need via Groups.Io<sailtamarack@...> wrote: On Fri, Jan 17, 2020 at 09:05 AM, Larry Rothman wrote:

A number of forum members have 2 or even 3 of the Nano's and I for one have
already modified mine to reduce noise and increase the freq range specs.
my $0.02

Larry,

Would you consider making a separate post detailing what you did to increase the freq range and reduce noise?

Roger

"Please note: Starting with NanoVNA-H rev3.4 D2 uses IN4148, if your NanoVNA
D2 is a Schottky diode, use the "vbat_offset 150" command to calibrate the
battery voltage."

So it made me wonder if the rest of the software worked with earlier H
versions

The D2 story is: few nanoVNA were shipped with any D2 installed.
1N4148 was installed by some, Schottky by others.
Just as most nanoVNA (except gecko) have type C USB connectors
while few have the resistors necessary for type C connection.
The only significance of D2 is for accuracy of the battery gauge,
which can be corrected using that "vbat" USB shell command
in any firmware which supports the battery gauge icon.

Hi Robert -

The answer will come

This comment in your first message is a concern:

I've tried creating the tmp folder and manually moving setup.py . It really didn't make much difference.

Ubuntu 18.04.3 LTS surely has a tmp folder for updates...

Linux distros generally have packaging schemes which typically do not use pip.
To the extent possible, a distro user should use the same package manager
to install python packages that was used to install python itself.

If python were manually installed from e.g. a tarball,
then the distro's package manager would not be implicated, and pip would be appropriate.

Using pip can be compared to crossing wires:


"In most cases, you should use pip within a virtual environment only."

On Sun, Jan 19, 2020 at 11:43 AM, Oristo wrote:

If you actually bothered to check the source, namely

then you would see that Hugen released separate firmware versions for -H4 (AKA
3.4) and -H (earlier) hardware

I did check and saw this comment by Hugen...

"Please note: Starting with NanoVNA-H rev3.4 D2 uses IN4148, if your NanoVNA D2 is a Schottky diode, use the "vbat_offset 150" command to calibrate the battery voltage."

So it made me wonder if the rest of the software worked with earlier H versions

But thanks anyway for your kind response....

Roger

Rudi:

I'll look into PyQt5 not installing issue. Rather be working on this than blowing snow in my driveway......

I've been running Ubuntu for years, and still I'm a novice. Thanks for taking time for the suggestion. The answer will come..

Best 73
Bob

On Sun, Jan 19, 2020 at 09:21 PM, Robert Rogers wrote:

Tried that as well. Results below

Hello Robert,

Too bad.
It helped me, therefore I told you.

What is strange for me, PyQt5 was downloaded OK, but was not installed.

73, Rudi DL5FA

I was thinking ahead and picked up an USB extender. Didn't work.

This summer I'm hoping to put up a modified .64 wavelength antenna (formerly on 11m).
Measuring the Z of the antenna then designing the feed matching network.

To keep any external influences out of the measurements, I thought it would be a good idea to take measurements from the ground.

Suggestions? An Amazon $7 solution didn't cut it.

73
Bob

Rudy:

Tried that as well. Results below

------------------------------------------------------------------------------------------

-OMEN-by-HP-Laptop:~/nanovna-saver$ sudo python3.7 -m pip install .
[sudo] password for mm:
The directory '/home/mm/.cache/pip/http' or its parent directory is not owned by the current user and the cache has been disabled. Please check the permissions and owner of that directory. If executing pip with sudo, you may want sudo's -H flag.
The directory '/home/mm/.cache/pip' or its parent directory is not owned by the current user and caching wheels has been disabled. check the permissions and owner of that directory. If executing pip with sudo, you may want sudo's -H flag.
Processing /home/mm/nanovna-saver
Collecting PyQt5 (from NanoVNASaver==0.2.2)
Downloading (3.2MB)
100% |████████████████████████████████| 3.2MB 232kB/s
Complete output from command python setup.py egg_info:
Traceback (most recent call last):
File "<string>", line 1, in <module>
File "/usr/lib/python3.7/tokenize.py", line 447, in open
buffer = _builtin_open(filename, 'rb')
FileNotFoundError: [Errno 2] No such file or directory: '/tmp/pip-build-a8hpixj_/PyQt5/setup.py'

----------------------------------------
Command "python setup.py egg_info" failed with error code 1 in /tmp/pip-build-a8hpixj_/PyQt5/
mm@bob-OMEN-by-HP-Laptop:~/nanovna-saver$ python3.7 -m pip install .
Processing /home/mm/nanovna-saver
Collecting PyQt5 (from NanoVNASaver==0.2.2)
Using cached
Complete output from command python setup.py egg_info:
Traceback (most recent call last):
File "<string>", line 1, in <module>
File "/usr/lib/python3.7/tokenize.py", line 447, in open
buffer = _builtin_open(filename, 'rb')
FileNotFoundError: [Errno 2] No such file or directory: '/tmp/pip-build-7clot2r6/PyQt5/setup.py'

----------------------------------------
Command "python setup.py egg_info" failed with error code 1 in /tmp/pip-build-7clot2r6/PyQt5/
-----------------------------------------------------------------------------------------------------------------------------------------------------
Before I forget tnx all for the help everyone

73

On Sun, Jan 19, 2020 at 04:06 AM, Rune Broberg wrote:

python3.7 nanovna-saver.py

tried that (btw tnx) results below
OMEN-by-HP-Laptop:~/nanovna-saver$ python3.7 nanovna-saver.py
Traceback (most recent call last):
File "nanovna-saver.py", line 17, in <module>
from NanoVNASaver.__main__ import main
File "/home/mm/nanovna-saver/NanoVNASaver/__main__.py", line 21, in <module>
from PyQt5 import QtWidgets, QtCore
ImportError: cannot import name 'QtWidgets' from 'PyQt5' (/usr/lib/python3/dist-packages/PyQt5/__init__.py)

You have to schort R13 and change? R22 upto included R24 also :

R22 and R23= 51 Ohms , R24 91 Ohms, R25 = 11 Ohms.

After this you have updated version 3.3 to version 3.4 as far as possible (without the power supply)

Jos.


Op 19-1-2020 om 19:45 schref Andy UA3RAW: